1. Field of the Invention
The present invention relates to a frequency modulation apparatus and a frequency modulation method for generating an image clock that is used for turning on/off a laser beam that scans an image bearing member, such as a photosensitive drum.
2. Related Background Art
In an electrophotographic image forming apparatus, generally, latent image forming is performed by turning on or off a laser beam emitted by a semiconductor laser and by exposing a photosensitive drum to the laser beam using a polygon mirror, and the image forming is performed by developing the latent image to obtain a toner image and transferring the toner image to a recording medium.
For this image forming apparatus, a constant clock is always employed as an image clock that is required for a laser controller that turns on or off a laser beam in accordance with input image data, and as a reference clock that is used as a reference for the generation of the image clock. The reason for this is as follows. If a reference clock is not constant, an image clock having a correct frequency can not be generated, a fluctuation is caused at this frequency, and the ON/OFF timing for the laser beam is shifted from the proper timing. Accordingly, the dot formation location for a latent image formed on a photosensitive member is slightly changed, and as a result, image distortion, misregistration and uneven coloring occur.
Further, for the image forming apparatus, a f-θ lens 40 is located between a polygon mirror 38 and a photosensitive member 42 in FIG. 1. The f-θ lens 40 possesses such optical characteristics as a laser beam condensing function and a distortion aberration correction function for ensuring linearity is maintained for scanning along a time axis, and is provided in order to scan a photosensitive member with a laser beam at a uniform speed. Therefore, the characteristics of the f-θ lens 40 greatly affect the printing accuracy in the scan direction.
In FIG. 10, a relationship between a print position and the distortion rate of the f-θ lens 40. The f-θ lens 40 has the f-θ characteristic, which is an optical characteristic shown by a curve in FIG. 10, and generally, as it is closer to the end area, the speed of scanning the photosensitive member 42 is increased, so that scanning at a completely uniform velocity is not obtained. That is, the distortion rate is increased from the center of the f-θ lens 40 (the central print position) to the end, and this is greatly related to a shift in the print positions at both ends of an image.
A detailed explanation for this problem will be described while referring to FIG. 11. When a specific pixel at the main scan end is denoted by Ps(N−1) and the next pixel is denoted by PsN, an interval Ds between the pixels at the end areas is represented asDs=PsN−Ps(N−1).Similarly, when a specific pixel in the scan center area is denoted by Pc(N−1) and the next pixel is denoted by PcN, a distance Dc between these pixels is represented asDc=PcN−Pc(N−1).Because of the above described characteristic of the f-θ lens 40, Ds>Dc is obtained, i.e., the pixel interval differs depending on the scanning position. As a result, an image is printed while the magnification rate differs, depending on the portions of the image, and an accurate image reproduction is not possible.
In order to minimize the print position shift that occurs due to the characteristic of the f-θ lens, conventionally, a frequency modulation technique is employed to modulate the frequency of an image write clock, and the shift in the print position is corrected electrically. There are, for example, a method for uniquely changing a frequency for one scan interval and a method for dividing one scan interval, and for modulating a frequency in an analog manner (e.g., Japanese Patent Application Laid-Open No. H2-282763).
However, as is apparent from FIGS. 10 and 11, the characteristic of the f-θ lens is complicated, and the distortion rate is increased, depending on the lens material. Therefore, according to the method used for uniquely changing the frequency for one scan interval, and the method for dividing one scan interval and modulating the frequency in an analog manner, accuracy can not be expected for the correction of the print position shift that occurs due to the characteristic of the f-θ lens. As a result, the printing quality is deteriorated.
For a color image forming apparatus, the above described f-θ lens 40 is provided for each of the colors Y, M, C and K. Because of the variations in the characteristics of the individual colors, the locations at which the photosensitive drum 42 is irradiated is shifted, even for pixels at the same position. As a result, misregistration in image forming occurs, and the image quality is remarkably deteriorated.
To resolve this problem, a well known apparatus is provided in Japanese Patent Application Laid-Open No. H9-218370 (Fuji Film). This apparatus partially modulates the frequency of an image clock along one line and ensures that scanning is performed at a uniform speed, so that any scan speed fluctuation produced by the f-θ characteristic is canceled.
However, as is shown in FIG. 15, even for the same lens, the f-θ characteristic differs for each apparatus due to errors in the size and attachment of the lens, and irradiation location shifts can not be avoided. Expensive lenses are required to reduce the variations in manufacturing lenses, and complicated operations are required to accurately position and attach the lenses.
A method for correcting variations in individual apparatuses is disclosed in Japanese Patent Application Laid-Open No. H11-198435 (Fuji Xerox). According to this method, instead of identical frequencies being modulated for a plurality of apparatuses, in each apparatus a registration mark is detected at multiple predetermined locations in the main scan direction, and frequency modulation, based on the detection of the distance shifted, is performed to correct for the shifting. According to this method, corrections in consonance with the characteristics of the individual apparatuses are enabled and performed.
However, as is shown in FIG. 12, since the absolute scan position for a specific pixel N is the cumulative accumulation from the scan start position, the positions of pixels 0 to (N−1) must be established, i.e., the pixel clock frequencies up to fn−1 must be determined. Therefore, all the frequency setup values must be used to perform calculations for the individual apparatuses for which readjustments are required. And since for these calculations, complicated algorithms and procedures are required, and since all the setup values must be stored in FIG. 13, an increased memory (RAM) capacity is required.
Further, as is shown in FIG. 25, the frequency modulation configuration of an image forming apparatus constituted by multi-beam laser comprises: a plurality of setting registers 72, 74 and 76 and a plurality of frequency modulating devices 71, 73 and 75 for generating image signal clocks 77, 78 and 79 for a plurality of laser beams. The setting registers 72, 74 and 76 hold setup values (variable-magnification coefficients) for one line in the main scan direction of a laser beam, or for the number of segments constituting the line. The frequency modulating devices 71, 73 and 75 generate the image signal clocks 77, 78 and 79 based on a reference clock signal Refclk, which is generated by a reference clock generating unit 70, and the setup values that are received from the corresponding setting registers 72, 74 and 76.
An image forming apparatus constituted by multi-beam lasers requires setting registers, i.e., correction tables, equivalent in number to the number of laser beams in order to perform corrections consonant with the characteristics of the f-θ lens. That is, a plurality of correction tables must be prepared in accordance with the locations passed by the laser beams and the characteristics of the f-θ lens. Therefore, for a plurality of laser beams, it takes time to perform corrections such as the preparation of correction tables, and the operation is very demanding.